VIRUSES AS MOLECULES 



TABLE I 



FRACTION REMAINING ABOVE BARRIER AFTER CBNTRiyUGATION 



1. Protein determined optically .26 



2. Protein determined by chemical analysis .26 



3. Virus infectlTlty .21 + .045 



It can be seen that the aunount of protein remaining in the top compartment 

 after the centrifuge was stopped was 26% of the original. It can also be seen 

 that the virus Infectlvlty remaining in the upper compartment was 21J^ of the 

 original, with a probable error of A-,^%. This value, 21 + A.JJf means that the 

 chances are even that the correct value of the infectlvlty lies within the range 

 21 + 4.5/f. It is evident that the percentage of virus infectlvlty remaining a- 

 bove is the same as the percentage of protein remaining above, within the probable 

 error of the determination. 



In order to appreciate the full significance of this result, it is neces- 

 sary to look at it from a slightly different point of view. It is possible to 

 calculate the sedimentation constant of the infectious principle from the amount 

 of infectlvlty remaining in the upper compartment. From the value of 21 + A-,^% 

 for the ratio of top compartment activity to original activity, one can calculate 

 a sedimentation constant of 178 + 11 x 10"^3 for the infectious principle. This 

 may be compared with the value, I65 x 10"^3, calculated from the optical data 

 for the protein in this same experiment. The lower value is seen to be in ex- 

 cellent agreement with the sedimentation constant of the virus protein. This 

 is really a fairly precise quantitative correlation of a biological function 

 with a chemical entity. It is probably the most precise quantitative test for 

 the identity of biological and chemical entitles yet applied to tobacco mosaic 

 virus. This result, taken in conjunction with the evidence previously present- 

 ed, affords a very strong Justification for making the assumption that the in- 

 fectious principle of tobacco mosaic is in reality firmly associated with the 

 nucleoprotein particles. 



A second important question concerns the size and shape of such an unusual 

 material as tobacco mosaic virus protein, the size and shape of the ultimate 

 particle or molecule. It has already been shown that the virus protein parti- 

 cles are large enough to be sedimented rapidly in a high speed centrifuge. 

 Therefore, the particles must be considerably larger than the molecules of 

 ordinary proteins like egg albumin. But even before anyone knew how big the 

 virus particles were, even before Stanley had isolated them and demonstrated 

 their protein nature, Takahashi and Rawlins had figured out from stream double 

 refraction studies that they were rod-like in shape. After the shape of to- 

 bacco mosaic virus was established, it became possible to compute, from the 

 data of filtration, diffusion and centrifugation experiments and from other in- 

 direct physicochemioal measurements, that tobacco mosaic virus particles are 

 rod-like bodies about 12-15 mu thick and several hundred mu long, with a mole- 

 cular weight of about 30 million. All of these computations were actually made 

 before smy one had ever seen a tobacco mosaic virus particle. 



After the initial phases of these studies were completed, the electron 

 microscope was perfected. This instrument uses electrons instead of light and 

 electromagnets Instead of lenses. Since a stream of electrons behaves like a 

 beam of light with very short wave length, the lower limit of resolution with 

 this instrijment is far beyond that with ordinary light. An electron micro- 

 graph of tobacco mosaic virus particles is reproduced in Figure 5* 



